GMP 4.2 tests failure newbie question

[Sergey Kashyrin]

Gentlemen,

I have a question regarding the following tests which failed after porting GMP 4.2 to 390 z/OS (not Linux not ASCII and not IEEE 
floating point !! so I can understand why "mpz_set_d wrong on 2^53+1")

The question is do you see anything really critical so I should worry about those failures ?

Regards,
Sergey

--------------------
MPN: 1 test failed
mpn_get_d wrong on random data
   sign     -1
   n        =0xC6843F29A331B7
   nsize    2
   exp      -9
   want     =[CC 63 42 1F 94 D1 98 D8] -109135648838040.8
   got      =[CC 63 42 1F 94 D1 98 DB] -109135648838040.9
CEE5207E The signal SIGABRT was received.
FAIL: t-get_d

--------------------
MPQ: 2 tests failed

CEE3212S The system detected an exponent-overflow exception (System Completion Code=0CC).
         From entry point __gmpn_get_d at compile unit offset +000005B2 at entry offset +000005B2 at address 1204986A.
FAIL: t-get_d

mpq_get_str wrong (passing NULL)
  base -16
  got  "FFFFFFFF"
  want "ffffffff"
  q=0xFFFFFFFF
CEE5207E The signal SIGABRT was received.
FAIL: t-get_str

--------------------
MPZ: 3 tests failed

CEE3212S The system detected an exponent-overflow exception (System Completion Code=0CC).
         From entry point check_onebit at compile unit offset +0000011E at entry offset +0000011E at address 11B7422E.
FAIL: t-get_d

mpz_set_d wrong on 2^53+1
  d =[4E 20 00 00 00 00 00 01] 9007199254740993
  got  =9007199254740992
  want =9007199254740993
CEE5207E The signal SIGABRT was received.
FAIL: t-set_d

CEE3212S The system detected an exponent-overflow exception (System Completion Code=0CC).
         From entry point __gmpn_get_d at compile unit offset +00000566 at entry offset +00000566 at address 1204981E.
FAIL: t-cmp_d